danvics/soulsync
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 1

Fix Picard albumartist orphan false positives

Browse source 
Teach the orphan file detector to match tracked files by both track artist and album artist. This prevents Picard-style albumartist/album (year)/track layouts from being reported as orphans when the DB track artist differs from the album artist.

Also check file albumartist tags during fallback matching and add a regression test for the reported Picard folder layout.
This commit is contained in:
Broque Thomas 2026-05-22 08:43:57 -07:00
parent 4179926899
commit 5d1f3c1b48
2 changed files with 116 additions and 12 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

47
core/repair_jobs/orphan_file_detector.py
View file

@ -67,22 +67,28 @@ class OrphanFileDetectorJob(RepairJob):
suffix = '/'.join(parts[-depth:]).lower()
known_suffixes.add(suffix)
# Build title+artist sets for tag-based fallback matching
# Build title+artist sets for fallback matching. Include both
# track artist and album artist so Picard-style albumartist paths
# don't look orphaned when tracks have featured/guest artists.
cursor.execute("""
SELECT t.title, ar.name FROM tracks t
LEFT JOIN artists ar ON ar.id = t.artist_id
SELECT t.title, track_ar.name, album_ar.name FROM tracks t
LEFT JOIN artists track_ar ON track_ar.id = t.artist_id
LEFT JOIN albums al ON al.id = t.album_id
LEFT JOIN artists album_ar ON album_ar.id = al.artist_id
WHERE t.title IS NOT NULL AND t.title != ''
""")
for row in cursor.fetchall():
title = (row[0] or '').lower().strip()
artist = (row[1] or '').lower().strip()
if title:
known_titles.add((title, artist))
# Also store normalized version (stripped of feat., parentheticals, etc.)
clean_t = _strip_extras(title)
clean_a = _strip_extras(artist)
if clean_t:
known_titles_clean.add((clean_t, clean_a))
for artist_value in (row[1], row[2]):
artist = (artist_value or '').lower().strip()
known_titles.add((title, artist))
# Also store normalized version (stripped of feat.,
# parentheticals, etc.)
clean_t = _strip_extras(title)
clean_a = _strip_extras(artist)
if clean_t:
known_titles_clean.add((clean_t, clean_a))
except Exception as e:
logger.error("Error reading known file paths from DB: %s", e, exc_info=True)
result.errors += 1
@ -144,14 +150,21 @@ class OrphanFileDetectorJob(RepairJob):
if audio:
file_title = ((audio.get('title') or [None])[0] or '').lower().strip()
file_artist = ((audio.get('artist') or [None])[0] or '').lower().strip()
file_albumartist = (
(audio.get('albumartist') or audio.get('album_artist') or [None])[0]
or ''
).lower().strip()
if file_title:
file_artists = [a for a in (file_artist, file_albumartist) if a]
if not file_artists:
file_artists = ['']
# Exact match first (fast path)
if (file_title, file_artist) in known_titles:
if any((file_title, artist) in known_titles for artist in file_artists):
is_known = True
else:
# Normalized match: strip (feat. X), [FLAC 16bit], etc.
clean_title = _strip_extras(file_title)
clean_artist = _strip_extras(file_artist)
clean_artist = _strip_extras(file_artists[0])
# Also try first artist only (handles "Gorillaz, Dennis Hopper" → "Gorillaz")
first_artist = clean_artist.split(',')[0].strip() if clean_artist else ''
if clean_title and (
@ -159,6 +172,16 @@ class OrphanFileDetectorJob(RepairJob):
(first_artist and (clean_title, first_artist) in known_titles_clean)
):
is_known = True
if clean_title and not is_known:
for artist in file_artists[1:]:
clean_artist = _strip_extras(artist)
first_artist = clean_artist.split(',')[0].strip() if clean_artist else ''
if (
(clean_title, clean_artist) in known_titles_clean or
(first_artist and (clean_title, first_artist) in known_titles_clean)
):
is_known = True
break
except Exception as e:
logger.debug("tag-based orphan check: %s", e)

81
tests/test_orphan_file_detector.py Normal file
View file

@ -0,0 +1,81 @@
"""Regression tests for the orphan file detector."""
from __future__ import annotations
import sqlite3
from pathlib import Path
from core.repair_jobs.base import JobContext
from core.repair_jobs.orphan_file_detector import OrphanFileDetectorJob
class _DB:
def __init__(self, path: Path) -> None:
self.path = path
def _get_connection(self):
return sqlite3.connect(self.path)
def _seed_library(db_path: Path) -> None:
conn = sqlite3.connect(db_path)
try:
conn.executescript(
"""
CREATE TABLE artists (
id INTEGER PRIMARY KEY,
name TEXT NOT NULL
);
CREATE TABLE albums (
id INTEGER PRIMARY KEY,
artist_id INTEGER NOT NULL,
title TEXT NOT NULL
);
CREATE TABLE tracks (
id INTEGER PRIMARY KEY,
album_id INTEGER NOT NULL,
artist_id INTEGER NOT NULL,
title TEXT NOT NULL,
file_path TEXT
);
INSERT INTO artists (id, name) VALUES
(1, 'Clouddead89'),
(2, 'Featured Artist');
INSERT INTO albums (id, artist_id, title) VALUES
(10, 1, 'Perfect Match Error');
INSERT INTO tracks (id, album_id, artist_id, title, file_path) VALUES
(100, 10, 2, 'Perfect Match', '/old/prefix/elsewhere.mp3');
"""
)
conn.commit()
finally:
conn.close()
def test_orphan_detector_accepts_picard_albumartist_folder_match(tmp_path: Path) -> None:
"""Picard paths use albumartist/album (year)/track - title.
Even when the DB track artist is a featured artist, the album artist
folder should be enough to recognize the file as tracked.
"""
db_path = tmp_path / "library.sqlite"
_seed_library(db_path)
transfer = tmp_path / "Clouddead89" / "Perfect Match Error (2026)"
transfer.mkdir(parents=True)
audio_path = transfer / "01 - Perfect Match.mp3"
audio_path.write_bytes(b"not a real mp3; filename fallback handles this")
findings = []
context = JobContext(
db=_DB(db_path),
transfer_folder=str(tmp_path),
config_manager=None,
create_finding=lambda **kwargs: findings.append(kwargs) or True,
)
result = OrphanFileDetectorJob().scan(context)
assert result.scanned == 1
assert result.findings_created == 0
assert findings == []